System and method for molecular diagnosis of depression based on boosting classification

Abstract

A method for diagnosing depression includes providing surface-enhanced laser desorption / ionization mass spectrometric (SELDI-MS) data of a plurality of proteins, said data obtained from a patient and comprising a plurality of peak values, and analyzing said peak values with an alternating decision tree comprising a set of tests of said data peaks values and associated prediction values, wherein said data is predictive of depression if a sum of the prediction values of said tree is greater than 1.0.

Description

CROSS REFERENCE TO RELATED UNITED STATES APPLICATIONS[0001]This application claims priority from “Molecular Diagnosis in Depression and Treatment based on Boosting Classification”, U.S. Provisional Application No. 60 / 644,141 of Wang, et al., filed Jan. 14, 2005, the contents of which are incorporated herein by reference.TECHNICAL FIELD[0002]This invention is directed to the automatic diagnosis of depression based on boosting classification of proteomic data.DISCUSSION OF THE RELATED ART[0003]Depression is one of the most common psychological problems in the states. Each year over 17,000,000 American adults experience a period of clinical depression. The cost in human suffering cannot be estimated. Depression can interfere with normal functioning, and frequently causes problems with work, social and family adjustment. Many people experiencing depression are misdiagnosed with physical illnesses instead. This is unfortunate because, with proper treatment, nearly 80% of those with depre...

AI Technical Summary

Benefits of technology

[0008]Exemplary embodiments of the invention as described herein generally include methods and systems for a boosting-based diagnosis method, on SELDI mass spectrometer data of depression patients and neurological controls. A molecular diagnosis methodology according to an embodiment of the invention uses boosting classification with IMAC SELDI protein profiling to provide an effective means to distinguish depression patients from the controls, which is confirmed by 10-fold cross validation and ROC analysis. The resulting decision rules are often simple and easy to interpret. These rules can be expressed in a simple and straightforward alternating decision tree format or in a conditional decision rule format. The decision rules identify the most discriminate features in SELDI MS data, which can be exploited as biomarkers for diagnosis.

[0009]A boosting based method according to an embodiment of the invention provides an efficient and objective method for physicians and psychologists to make a diagnosis for patients, especially those who could be misdiagnosed with other diseases. These methods provide an efficient technique for database-guided diagnosis of depression and other neurological diseases.

Problems solved by technology

The cost in human suffering cannot be estimated.

Depression can interfere with normal functioning, and frequently causes problems with work, social and family adjustment.

Images